Original article

C-reactive protein trajectory to predict colorectal anastomotic

leak: PREDICT Study
B. D. Stephensen1 , F. Reid1 , S. Shaikh4,5 , R. Carroll1,2 , S. R. Smith1,2,3 and P. Pockney1,2,3 , on behalf
of the PREDICT Study Group collaborators*
1
Department of Colorectal Surgery and 2 Hunter Surgical Clinical Research Unit, John Hunter Hospital, and 3 School of Medicine and Public Health,
University of Newcastle, Newcastle, New South Wales, Australia, and 4 Department of Surgery, Aberdeen Royal Infirmary and 5 Department of Surgery,
University of Aberdeen, Aberdeen, UK

Downloaded from https://academic.oup.com/bjs/article/107/13/1832/6139506 by guest on 05 September 2022

Correspondence to: Dr B. D. Stephensen, Department of Colorectal Surgery, John Hunter Hospital, New Lambton Heights, New South Wales 2305,
Australia (e-mail: [email protected])

Background: Anastomotic leak is a common complication after colorectal surgery, associated with
increased morbidity and mortality, and poorer long-term survival after oncological resections. Early
diagnosis improves short-term outcomes, and may translate into reduced cancer recurrence. Multiple
studies have attempted to identify biomarkers to enable earlier diagnosis of anastomotic leak. One study
demonstrated that the trajectory of C-reactive protein (CRP) levels was highly predictive of anastomotic
leak requiring intervention, with an area under the curve of 0⋅961. The aim of the present study was to
validate this finding externally.
Methods: This was a prospective international multicentre observational study of adults undergoing
elective colorectal resection with an anastomosis. CRP levels were measured before operation and for
5 days afterwards, or until day of discharge if earlier than this. The primary outcome was anastomotic
leak requiring operative or radiological intervention.
Results: Between March 2017 and July 2018, 933 patients were recruited from 20 hospitals across
Australia, New Zealand, England and Scotland. Some 833 patients had complete CRP data and were
included in the primary analysis, of whom 41 (4⋅9 per cent) developed an anastomotic leak. A change in
CRP level exceeding 50 mg/l between any two postoperative days had a sensitivity of 0⋅85 for detecting a
leak, and a high negative predictive value of 0⋅99 for ruling it out. A change in CRP concentration of more
than 50 mg/l between either days 3 and 4 or days 4 and 5 after surgery had a high specificity of 0⋅96–0⋅97,
with positive likelihood ratios of 4⋅99–6⋅44 for a leak requiring intervention.
Conclusion: This study confirmed the value of CRP trajectory in accurately ruling out an anastomotic
leak after colorectal resection.
∗ Members of the PREDICT Study Group are co-authors of this article and can be found under the heading

Collaborators.
Presented to the Annual Scientific Congress of the Royal Australasian College of Surgeons, Bangkok, Thailand, May
2019, and the Annual Meeting of the Association of Coloproctology of Great Britain and Ireland, Dublin, Ireland, July
2019; published in abstract form as Colorectal Dis 2019; 21(Suppl 2): 4

Paper accepted 25 May 2020

Published online 16 July 2020 in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.11812

Introduction diagnosis, however, has the ability to improve short-term

morbidity and mortality, and may translate into reduced
Anastomotic leak is a common and feared complication of
cancer recurrence in the long term6 .
colorectal surgery, with reported rates between 2 and 14 per
cent1–5 . It is associated with increased morbidity and mor- Previous studies have attempted to identify biomarkers
tality, as well as worse oncological outcomes3–9 . Despite that may assist with early diagnosis of anastomotic leakage.
significant advances in surgical technique and perioperative C-reactive protein (CRP) appears to be one of the most
care, including enhanced recovery protocols, the incidence widely studied of these biomarkers. A meta-analysis11 of all
of anastomotic leak does not seem to be decreasing10 . Early studies performed on the use of CRP for early detection

© 2020 BJS Society Ltd BJS 2020; 107: 1832–1837

Published by John Wiley & Sons Ltd
C-reactive protein trajectory to predict colorectal anastomotic leak 1833

of anastomotic leak found that cut-off levels above certain anastomotic leak, and assessment of readiness for dis-
points on postoperative days 3, 4 and 5 had good diagnostic charge. This was a pragmatic observational study, and the
accuracy, with a pooled area under the curve (AUC) of decision to use mechanical or oral antibiotic bowel prepara-
around 0⋅80. tion and diverting stomas was based entirely on the treating
A recent study12 has suggested that the rate of change, or surgeon’s preferences. Clinicians were not blinded to the
trajectory, of CRP levels may be even more useful than a CRP results.
cut-off point alone. This study revealed that the trajectory CRP levels were measured before surgery and daily up to
of CRP was not only highly accurate in predicting leak day 5 after operation, or until day of discharge for patients
requiring intervention (AUC 0⋅961) but also helpful in discharged earlier than this. Both the trajectory of CRP and
excluding it, with a negative predictive value of 99⋅3 per daily cut-off points for CRP were analysed prospectively

Downloaded from https://academic.oup.com/bjs/article/107/13/1832/6139506 by guest on 05 September 2022

cent12 . It may be that CRP trajectory cannot only predict with regard to their accuracy in prediction of anastomotic
leak more accurately, but that patients with a reassuring leakage. Anastomotic leak was defined as a defect in the
trajectory could be discharged earlier, which is of interest intestinal wall at the site of the anastomosis requiring oper-
in the era of enhanced recovery. The aim of this study was ative or radiological intervention. A secondary analysis was
to determine the external validity of both these findings in undertaken that considered all anastomotic leaks, includ-
a larger, multicentre cohort, and to validate the accuracy ing those managed medically. Comparison was performed
of increases in CRP concentration on days 3, 4 and 5 between leakage and a CRP trajectory of 50 mg/l per day12
described by Singh and colleagues11 . as well as the cut-off points identified in the meta-analysis
of Singh and colleagues11 .

Methods
Statistical analysis
This was a prospective international multicentre obser-
vational study of adults aged over 18 years undergoing The null hypothesis was that there is no association
elective or semiurgent colorectal anastomosis. Patients between CRP trajectory and diagnosis of anastomotic leak
having elective colorectal resection or restoration of bowel requiring intervention. To establish with 95 per cent con-
continuity with anastomosis (ileocolic, ileorectal, colocolic, fidence and 80 per cent power an AUC greater than 0⋅85
colorectal or coloanal), for any indication (benign or malig- (considered highly accurate), 740 patients were required
nant), were eligible for inclusion. Ileoanal (pouch) anasto- for analysis.
moses were excluded. Descriptive statistics were used to describe the main vari-
Multicentre Australian ethical approval was granted ables of interest, with continuous variables summarized
by the institutional ethical review board, Hunter New using mean(s.d.) or median (range) and categorical vari-
England Human Research Ethics Committee, New ables summarised using frequency and percentages.
South Wales, Australia, in February 2017 (HNEHREC After imputation of missing values, the predictive per-
17/02/15/4.03). In New Zealand, ethical approval was formance of an increase of more than 50 mg/l in CRP
granted by the Southern Health and Disability Ethics between any 2 consecutive days, as well as between each
Committee in February 2018 (17/STH/246). Centres pair of days individually, in predicting anastomotic leak was
in the UK proceeded with the study on the basis of local assessed using sensitivity, specificity, positive and negative
quality assurance audit approvals at each site. Patients were diagnostic likelihood ratios, and positive (PPV) and nega-
recruited from 20 participating hospitals across four coun- tive (NPV) predictive value.
tries, commencing in March 2017. Recruitment at each Separate logistic regression models were undertaken for
site was coordinated by a training subspecialty fellow or change in CRP level between each pair of consecutive days,
surgical registrar. Individual written informed consent was as well as the CRP score recorded on each day. For these
obtained from all patients in Australia and New Zealand, analyses, the change in CRP level and CRP concentration
as required by the relevant ethics approvals. on each day were analysed as continuous variables (no
Prospective data collection included: baseline demo- predefined cut-off point was used). The AUC was then
graphics, ASA fitness grade, BMI, co-morbidities, receipt calculated for each model.
of neoadjuvant chemotherapy or radiotherapy, procedure Statistical analysis was performed by Clinical Research
performed, level of anastomosis, postoperative complica- Design, IT and Statistical Support, Hunter Medical
tions, duration of hospital stay and anastomotic leakage. Research Institute, using SAS® version 9.4 (SAS Institute,
All clinical management was done at the treating teams’ Cary, North Carolina, USA) and R (R Foundation for
discretion, including investigation and management of Statistical Computing, Vienna, Austria).

© 2020 BJS Society Ltd www.bjs.co.uk BJS 2020; 107: 1832–1837

Published by John Wiley & Sons Ltd
1834 B. D. Stephensen, F. Reid, S. Shaikh, R. Carroll, S. R. Smith and P. Pockney

Table 1 Patient characteristics Table 3 Leak rate by level of anastomosis

Anastomotic Anastomotic
No leak leak No leak leak
(n = 792) (n = 41) P‡
Ileocolic anastomosis 344 9 (2⋅5)
Age (years)* 64(15) 64(12) 0⋅934 Colocolic anastomosis 96 6 (5⋅9)
Sex < 0⋅001 High anterior resection 197 11 (5⋅3)
F 372 8 (2⋅1) Low anterior resection 45 4 (8)
M 420 33 (7⋅3) Low anterior resection + stoma 110 11 (9⋅1)
BMI (kg/m2 )† 28(6) 28(6) 0⋅461§
Total 792 41 (4⋅9)
Smoker 115 8 (6⋅5) 0⋅368

Downloaded from https://academic.oup.com/bjs/article/107/13/1832/6139506 by guest on 05 September 2022

Diabetic 115 10 (8⋅0) 0⋅112 Values in parentheses are percentages.
Duration of hospital stay (days)† 6 (2–166) 19 (3–76)
Death 4 (0⋅5) 0 (0)
Table 4 Diagnostic indices for ability of daily C-reactive protein
Values in parentheses are percentages unless indicated otherwise: *values change exceeding 50 mg/l to predict anastomotic leak requiring
are mean(s.d.) and †median (range). ‡Fishers exact test except §Student’s intervention
t test.
Timing of CRP
increase > 50 mg/l Sensitivity Specificity PLR NLR PPV NPV

Between any 2 days 0⋅85 0⋅51 1⋅75 0⋅29 0⋅08 0⋅99

Table 2 Indication for operation From day 1 to day 2 0⋅63 0⋅57 1⋅48 0⋅64 0⋅07 0⋅97
Anastomotic From day 2 to day 3 0⋅32 0⋅90 3⋅10 0⋅76 0⋅14 0⋅96
No leak leak From day 3 to day 4 0⋅20 0⋅96 4⋅99 0⋅84 0⋅21 0⋅96

Colonic cancer 405 17 (4⋅0) From day 4 to day 5 0⋅17 0⋅97 6⋅44 0⋅85 0⋅25 0⋅96

Rectal cancer 147 15 (9⋅3) CRP, C-reactive protein; PLR, positive likelihood ratio; NPR, negative
Diverticular disease 34 3 (8) likelihood ratio; PPV, positive predictive value; NPV, negative predictive
Restoration of bowel continuity 78 2 (3) value.
Inflammatory bowel disease 52 1 (2)
Other 76 3 (4)
Table 5 Area under the curve for assessing accuracy of
Total 792 41 (4⋅9) C-reactive protein trajectory and daily C-reactive protein
accuracy in predicting anastomotic leak requiring intervention
Values in parentheses are percentages.
AUC

Change in CRP > 50 mg/l

Results
From day 1 to day 2 0⋅65
From day 2 to day 3 0⋅61
Between March 2017 and July 2018, 933 patients were
From day 3 to day 4 0⋅57
recruited from 20 hospitals across Australia, New Zealand,
From day 4 to day 5 0⋅52
England and Scotland. Some 833 patients had complete
Daily CRP accuracy
data and were included in the primary analysis. Follow-up Day 1 0⋅56
was complete to 30 days for all patients. Day 2 0⋅64
In total, 100 patients were excluded from the analysis Day 3 0⋅67
owing to incomplete CRP data. Of 40 patients discharged Day 4 0⋅74
before day 3 blood tests had been carried out, CRP levels Day 5 0⋅79
for days 1 and 2 were available for 26, and these patients
AUC, area under the curve; CRP, C-reactive protein.
were included in the analysis. There were no readmissions
for anastomotic leak in patients discharged on or before
day 3. Of the 41 patients who had an anastomotic leak requiring
The overall leak rate in this study was 4⋅9 per cent (41 of intervention, 37 had at least one operative intervention; 32
833 patients). Demographic and clinical characteristics of patients underwent reoperation, three patients had rectal
the study population are shown in Table 1. Anastomotic leak drainage alone, and two patients required percutaneous
occurred more often in men than women (7⋅3 versus 2⋅1 per drainage in addition to rectal drainage. Four leaks required
cent; P < 0⋅001). Patients who developed an anastomotic radiological intervention (percutaneous drainage) alone.
leak had a longer hospital stay (median 19 versus 6 days), A further 21 patients had an anastomotic leak identified
but there was no death associated with leak in this study. that was managed medically; in accordance with the study

© 2020 BJS Society Ltd www.bjs.co.uk BJS 2020; 107: 1832–1837

Published by John Wiley & Sons Ltd
C-reactive protein trajectory to predict colorectal anastomotic leak 1835

Table 6 Comparison of C-reactive protein trajectory and validated cut-off points for predicting anastomotic leak requiring intervention

Sensitivity Specificity PLR NLR PPV NPV

CRP increase > 50 mg/l

For 2 consecutive days 0⋅29 0⋅92 3⋅77 0⋅77 0⋅16 0⋅96
For 3 consecutive days 0⋅05 0⋅99 6⋅60 0⋅96 0⋅25 0⋅95
From day 1 to day 2, and day 2 to day 3 0⋅22 0⋅93 3⋅23 0⋅84 0⋅14 0⋅96
From day 2 to day 3, and day 3 to day 4 0⋅10 0⋅99 7⋅92 0⋅91 0⋅29 0⋅95
From day 3 to day 4, and day 4 to day 5 0⋅05 0⋅99 9⋅85 0⋅96 0⋅33 0⋅95
From day 1 to day 2, day 2 to day 3, and day 3 to day 4 0⋅05 0⋅99 6⋅60 0⋅96 0⋅25 0⋅95

Downloaded from https://academic.oup.com/bjs/article/107/13/1832/6139506 by guest on 05 September 2022

From day 2 to day 3, day 3 to day 4, and day 4 to day 5 0⋅02 1⋅00 15⋅84 0⋅98 0⋅50 0⋅95
CRP cut-off value
> 172 mg/l on day 3 0⋅46 0⋅74 1⋅77 0⋅73 0⋅08 0⋅96
> 124 mg/l on day 4 0⋅68 0⋅69 2⋅16 0⋅46 0⋅10 0⋅98
> 144 mg/l on day 5 0⋅51 0⋅84 3⋅21 0⋅58 0⋅14 0⋅97

CRP, C-reactive protein; PLR, positive likelihood ratio; NLR, negative likelihood ratio; PPV, positive predictive value; NPV, negative predictive value.

definition of leak (leakage requiring intervention), these Table 7 Diagnostic indices for ability of daily C-reactive protein
patients were included in the no-leak group for the primary change exceeding 50 mg/l to predict all leaks, including those
analyses. managed medically

Leak rates are shown by indication for surgery and level Timing of CRP
increase > 50 mg/l Sensitivity Specificity PLR NLR PPV NPV
of anastomosis in Tables 2 and 3 respectively. More than
two-thirds of the cohort underwent surgery for colonic or Between any 2 days 0⋅87 0⋅52 1⋅82 0⋅25 0⋅13 0⋅98

rectal cancer. An ileocolic anastomosis was the most fre- From day 1 to day 2 0⋅69 0⋅58 1⋅66 0⋅53 0⋅12 0⋅96
From day 2 to day 3 0⋅35 0⋅91 3⋅80 0⋅71 0⋅23 0⋅95
quently performed, which had the lowest leak rate at 2⋅5
From day 3 to day 4 0⋅16 0⋅96 4⋅29 0⋅87 0⋅26 0⋅93
per cent. A χ2 test of independence revealed a significant
From day 4 to day 5 0⋅13 0⋅97 4⋅97 0⋅89 0⋅29 0⋅93
relationship between the level of anastomosis and leakage
(χ2 = 10⋅10, 1 d.f., P = 0⋅039; n = 833); the lower the anas- CRP, C-reactive protein; PLR, positive likelihood ratio; NLR, negative
likelihood ratio; PPV, positive predictive value; NPV, negative predictive
tomosis, the higher the risk of leak. Overall, 121 patients value.
received a defunctioning stoma at the primary procedure.
There was no significant difference in the leak rates for low
forms of modelling ranged from 0⋅95 to 0⋅98 across all
anterior resection with and without a stoma (9⋅1 versus 8⋅2
evaluations.
per cent; P = 1⋅000).
A change in CRP level of more than 50 mg/l between
any two consecutive postoperative days had a sensitivity Secondary analysis
for predicting leak of 0⋅85, with a NPV of 0⋅99 (Table 4). To assess the value of the same CRP trajectory (over
The specificity of this change in CRP improved from 0⋅57 50 mg/l per day) in identifying all anastomotic leaks, rather
between days 1 and 2 after surgery to 0⋅96 between days than only those requiring intervention, a secondary analy-
3 and 4, and 0⋅97 between days 4 and 5. AUC values for sis was performed that included the 21 medically managed
the CRP trajectory exceeding 50 mg/l per day ranged from leaks in the anastomotic leak group, instead of the no-leak
0⋅52 to 0⋅65 (Table 5). group (Table 7). These results showed slightly improved
Considering daily cut-off points for predicting leak, the PPVs and marginally lower NPVs.
AUC for assessing accuracy ranged from 0⋅56 on day 1 to
0⋅79 on day 5 (Table 5). Trajectory modelling over mul-
Discussion
tiple consecutive pairs of days revealed specificity values
between 0⋅92 and 1⋅00, but with a sensitivities ranging This large prospective analysis of the accuracy of CRP
from 0⋅02 to 0⋅29, whereas analyses using the daily cut-off testing in diagnosing anastomotic leakage has shown that,
points identified by Singh and colleagues11 resulted in although CRP trajectory and cut-off points are not as
sensitivity values ranging from 0⋅46 to 0⋅68, with speci- accurate as expected when subjected to a large multicentre
ficities between 0⋅69 and 0⋅84 (Table 6). NPVs for both study, they certainly have value in diagnosing and excluding

© 2020 BJS Society Ltd www.bjs.co.uk BJS 2020; 107: 1832–1837

Published by John Wiley & Sons Ltd
1836 B. D. Stephensen, F. Reid, S. Shaikh, R. Carroll, S. R. Smith and P. Pockney

this significant surgical condition. It has also confirmed the in identifying all anastomotic leaks, rather than only those
speed of recruitment and educational transfer that occurs requiring intervention, showed slightly improved PPVs,
when resources from research collaboratives are combined. although all were still low, and marginally lower NPVs.
The present study failed to replicate the accuracy of CRP Thus, CRP trajectory seemed more accurate for anasto-
trajectory seen in the initial single-centre study. Although motic leaks requiring intervention, particularly for ruling
AUC values did not reach those considered to be highly out a leak.
predictive (0⋅85), there was evidence of value in CRP test- Another potential limitation of the study is bias regarding
ing. The utility of CRP trajectory appeared to be related to CRP as a biomarker. Clinicians were not blinded to the
its high early postoperative NPV as well as its high speci- daily CRP results, and may have acted on those results
ficity from day 3 onwards. For an increase in CRP con- even earlier than expected. It is apparent that patients who

Downloaded from https://academic.oup.com/bjs/article/107/13/1832/6139506 by guest on 05 September 2022

centration exceeding 50 mg/l between days 1 and 2, the developed anastomotic leakage were well managed in this
NPV was 0⋅97, meaning that patients whose CRP level multicentre study, as reflected by the mortality rate of zero
did not follow this trajectory after surgery had a 97 per following leakage. Intervention may have occurred earlier
cent chance of not developing an anastomotic leak. Early than it would if the clinicians had been blinded to the
specificity was confounded by false-positives (CRP level CRP measurements, resulting in the low mortality rate,
increasing owing to other sources of infection and surgical and poorer accuracy, sensitivity and PPV than in previous
trauma), but the specificity of 0⋅96–0⋅97 for a CRP increase studies.
exceeding 50 mg/l between days 3 and 4 or days 4 and 5 Although this study did not investigate cost-effectiveness,
also highlights the value of CRP monitoring in excluding CRP measurement is a cheap test (Australian $9⋅10,
anastomotic leakage for patients whose CRP level is not Pathology NSW, Newcastle, New South Wales, Australia;
increasing in the later stages of admission. Therefore, lack €5⋅43, exchange rate 29 May 2020). Samples are usually
of an increase in CRP level of over 50 mg/l in any 24-h collected in lithium heparin tubes of the type used for
period is an excellent facilitator of early discharge from electrolyte and renal function tests; these tests are com-
hospital. monly performed daily in the early postoperative course
Using cut-off points in this study appeared to provide after colorectal surgery, so CRP testing does not involve
improved sensitivity but lower specificity compared with extra stress or inconvenience for the patient. Despite the
trajectory testing. This makes implicit sense considering lower accuracy here compared with previously published
the nature of anastomotic leakage and CRP testing. A values, it would seem to be a useful routine intervention
high concentration in an isolated test should be more in the postoperative setting of colorectal anastomosis that
likely to diagnose a leak, whereas lack of an increase over can provide early reassurance and facilitate safe discharge.
consecutive days of testing should be more likely to rule
out a leak. To this extent, the concept of using both cut-off Collaborators
points and trajectory seems intuitive.
Members of the PREDICT Study Group are as follows (*fellow/trainee
One possible limitation of the study is the definition of
principal investigator). Study steering committee: B. Stephensen* (John
anastomotic leakage based on the requirement for opera- Hunter Hospital, Newcastle, New South Wales, Australia); P. Pockney,
tive or radiological intervention. This may have resulted S. R. Smith (University of Newcastle, Newcastle, New South Wales,
in a lower than expected leak rate (4⋅9 per cent), thus Australia); S. Shaikh* (St Mark’s Hospital, Harrow, UK); D. Morton,
affecting sensitivity and PPV. Broadening the definition of T. Pinkney (University of Birmingham, Birmingham, UK); J. Glaseby*,
R. Wilkin* (West Midlands Research Collaborative, Birmingham, UK);
anastomotic leak is challenging, however, and potentially
R. Carroll (John Hunter Hospital, Newcastle, New South Wales, Aus-
results in variable inclusion of patients owing to subjectivity tralia). Study writing committee: B. Stephensen*, S. R. Smith, P. Pock-
regarding what constitutes a leak. The International Study ney. Participating investigators and centres: B. Stephensen*, R. Carroll, P.
Group of Rectal Cancer13 proposed the widely used scale Pockney, F. Reid*, S. R. Smith (John Hunter Hospital, Newcastle, New
that grades anastomotic leak according to its impact on South Wales, Australia); M. Hong*, I. Faragher (Western Hospital, Mel-
bourne, Victoria, Australia); D. A. Clark, A. Edmundson, B. Stephensen*
clinical management. Grade B leaks require intervention (Royal Brisbane and Women’s Hospital, Brisbane, Queensland, Australia);
without recourse to the peritoneal cavity (such as percu- G. Armapatzis, S. Shaikh* (St Marks Hospital, Harrow, UK); C. Chal-
taneous drainage), whereas grade C leaks require reoper- land, M. Lee*, L. Sellors (Sheffield Teaching Hospitals, Sheffield, UK);
ation. Anastomotic leaks that require intervention are of T. Chittleborough*, T. Eglington, M. Falloon, D. Wright* (Christchurch
Hospital, Christchurch, New Zealand); M. Ali, D. Yeung* (Heartlands
most interest to the surgeon; however, the septic responses
Hospital, Birmingham, UK); D. Ashmore* (Hull and East Yorkshire Hos-
to a grade A leak, while managed conservatively, may pitals, Hull, UK); K. Knight*, H. Van Wyck (Glasgow Royal Infirmary,
still contribute to poorer long-term oncological outcomes. Glasgow, UK); J. McGovern, J. Park*, S. Rose (Queen Elizabeth Univer-
Analysis of the same CRP trajectory (over 50 mg/l per day) sity Hospital, Glasgow, UK); H. Cheung*, M. J. F. X. Rickard (Concord

© 2020 BJS Society Ltd www.bjs.co.uk BJS 2020; 107: 1832–1837

Published by John Wiley & Sons Ltd
C-reactive protein trajectory to predict colorectal anastomotic leak 1837

Hospital, Sydney, New South Wales, Australia); H. Cain, A. G. Heriot, S. generates prodigious use of hospital resources. Colorectal Dis
McKeown, F. Reid* (Peter McCallum Cancer Centre, Melbourne, Victo- 2009; 11: 917–920.
ria, Australia); A. Ball*, S. Ramcharan, A. Sinha (Warwick Hospital, War- 5 Bruce J, Krukowski ZH, Al-Khairy G, Russell EM,
wick, UK); M. Gregori* (Worcestershire Acute Hospitals, Worcester, UK);
Park KG. Systematic review of the definition of anastomotic
J. Glaseby*, J. Santos Torres, Y. Sinha (Queen Elizabeth Hospital, Birm-
leak after gastrointestinal surgery. Br J Surg 2001; 88:
ingham, UK); M. Al-Azzawi*, L. Clark, S. Shimu (Basingstoke and North
1157–1168.
Hampshire Hospital, Basingstoke, UK); N. Copertino*, D. A. Grieve, S.
Ryan (Sunshine Coast University Hospital, Sunshine Coast, Queensland, 6 Branagan G, Finnis D; Wessex Colorectal Cancer Audit
Australia); Q. Ain, J. Lahtela, R. Wilkin* (University Hospital Coventry, Working Group. Prognosis after anastomotic leakage in
Coventry, UK); E. Li*, J. Vatish (Walsall Manor Hospital, Walsall, UK); colorectal surgery. Dis Colon Rectum 2005; 48: 1021–1026.
C. J. Young, A. Zahid* (Royal Prince Alfred Hospital, Sydney, New South 7 Khan AA, Wheeler JM, Cunningham C, George B,
Wales, Australia). Kettlewell M, Mortensen NJ. The management and

Downloaded from https://academic.oup.com/bjs/article/107/13/1832/6139506 by guest on 05 September 2022

outcome of anastomotic leaks in colorectal surgery. Colorectal
Dis 2008; 10: 587–592.
Acknowledgements 8 Boccola MA, Buettner PG, Rozen WM, Siu SK,
Stevenson AR, Stitz R et al. Risk factors and outcomes
The study was funded partially by a grant from the Col-
for anastomotic leakage in colorectal surgery: a
orectal Surgical Society of Australia and New Zealand
single-institution analysis of 1576 patients. World J Surg
Foundation, a Royal Australasian College of Surgeons 2011; 35: 186–195.
Foundation for Surgery Small Project Grant, and a grant 9 Mirnezami A, Mirnezami R, Chandrakumaran K, Sasapu K,
from the John Hunter Charitable Trust. Sagar P, Finan P. Increased local recurrence and reduced
Disclosure The authors declare no conflict of interest. survival from colorectal cancer following anastomotic leak:
systematic review and meta-analysis. Ann Surg 2011; 253:
890–899.
References
10 Shogan BD, Carlisle EM, Alverdy JC, Umanskiy K. Do we
1 Hyman N, Manchester TL, Osler T, Burns B, Cataldo PA. really know why anastomoses leak? J Gastrointest Surg 2013;
Anastomotic leaks after intestinal anastomosis: it’s later than 17: 1698–1707.
you think. Ann Surg 2007; 245: 254–258. 11 Singh PP, Zeng IS, Srinivasa S, Lemanu DP, Connolly AB,
2 Ives A, Panis Y, Trancart D, Regimbeau JM, Pocard M, Hill AG. Systematic review and meta-analysis of use of serum
Valleur P. Factors associated with clinically significant C-reactive protein levels to predict anastomotic leak after
anastomotic leakage after large bowel resection: colorectal surgery. Br J Surg 2014; 10: 339–346.
multivariate analysis of 707 patients. World J Surg 2002; 26: 12 Smith SR, Pockney P, Holmes R, Doig F, Attia J, Holliday E
499–502. et al. Biomarkers and anastomotic leakage in colorectal
3 Phitayakorn R, Delaney CP, Reynolds HL, Champagne BJ, surgery: C-reactive protein trajectory is the gold standard.
Heriot AG, Neary P et al. Standardized algorithms for ANZ J Surg 2018; 88: 440–444.
management of anastomotic leaks and related abdominal and 13 Rahbari NN, Weitz J, Hohenberger W, Heald RJ, Moran B,
pelvic abscesses after colorectal surgery. World J Surg 2008; Ulrich A et al. Definition and grading of anastomotic leakage
32: 1147–1156. following anterior resection of the rectum: a proposal by the
4 Frye J, Bokey EL, Chapuis PH, Sinclair G, Dent OF. International Study Group of Rectal Cancer. Surgery 2010;
Anastomotic leakage after resection of colorectal cancer 147: 339–351.

© 2020 BJS Society Ltd www.bjs.co.uk BJS 2020; 107: 1832–1837

Published by John Wiley & Sons Ltd
 European Colorectal Congress
LORECTAL
CO
N

CO
EUROPEA

NGR SS
2022

E
N

d
28 November – 1 December 2022, St.Gallen, Switzerland
28

l
an
ov er
–1 tz
D ec wi
· S t. G all e n · S

Monday, 28 November 2022 Tuesday, 29 November 2022 Wednesday, 30 November 2022

09.50 9.00 9.00

Opening and welcome CONSULTANT‘S CORNER Advanced risk stratification in colorectal
Jochen Lange, St.Gallen, CH Michel Adamina, Winterthur, CH cancer – choosing wisely surgery and
adjuvant therapy
10.00 10.30 Philip Quirke, Leeds, UK
It is leaking! Approaches to salvaging an COFFEE BREAK
anastomosis 09.30
Willem Bemelman, Amsterdam, NL 11.00 Predictors for Postoperative Complications
SATELLITE SYMPOSIUM and Mortality
10.30 Ronan O‘Connell, Dublin, IE
Predictive and diagnostic markers
of anastomotic leak 10.00
Andre D‘Hoore, Leuven, BE 11.45 Segmental colectomy versus extended
Trends in colorectal oncology and colectomy for complex cancer
11.00 clinical insights for the near future Quentin Denost, Bordeaux, FR
SATELLITE SYMPOSIUM Rob Glynne-Jones, London, UK
10.30
12.15 COFFEE BREAK
LUNCH
11.45 11.00
Of microbes and men – the unspoken 13.45 Incidental cancer in polyp - completion
story of anastomotic leakage VIDEO SESSION surgery or endoscopy treatment alone?
James Kinross, London, UK Laura Beyer-Berjot, Marseille, FR
14.15
12.15 SATELLITE SYMPOSIUM 11.30
LUNCH SATELLITE SYMPOSIUM
13.45
Operative techniques to reduce 15.00
anastomotic recurrence in Crohn’s disease COFFEE BREAK 12.00
Laura Hancock, Manchester, UK Less is more – pushing the boundaries
15.30 of full-thickness rectal resection
14.15 The unsolved issue of TME: Xavier Serra-Aracil, Barcelona, ES
Innovative approaches in the treatment open, robotic, transanal, or laparoscopic –
of complex Crohn Diseases perianal fistula shining light on evidence and practice 12.30
Christianne Buskens, Amsterdam, NL Des Winter, Dublin, IE LUNCH
Jim Khan, London, UK
14.45 Brendan Moran, Basingstoke, UK 14.00
To divert or not to divert in Crohn surgery – Management of intestinal
technical aspects and patient factors 16.30 neuroendocrine neoplasia
Pär Myrelid, Linköping, SE SATELLITE SYMPOSIUM Frédéric Ris, Geneva, CH
15.15 14.30
COFFEE BREAK Poster Presentation & Best Poster Award
Michel Adamina, Winterthur, CH
15.45 17.15
Appendiceal neoplasia – when to opt for a Lars Pahlman lecture 15.00
minimal approach, when and how to go for Søren Laurberg, Aarhus, DK SATELLITE SYMPOSIUM
a maximal treatment
Tom Cecil, Basingstoke, Hampshire, UK

16.15 15.45
SATELLITE SYMPOSIUM COFFEE BREAK

16.15
Reoperative pelvic floor surgery –
17.00 er 2022 dealing with perineal hernia, reoperations,
Outcomes of modern induction therapies ecemb
ay, 1 D
and complex reconstructions
and Wait and Watch strategies, Hope or Hype h u rs d urgery
rectal S
T Guillaume Meurette, Nantes, FR
Antonino Spinelli, Milano, IT C o lo
class in y
Master logy Da 16.45
17.30 Procto Salvage strategies for rectal neoplasia
EAES Presidential Lecture - Use of ICG in Roel Hompes, Amsterdam, NL
colorectal surgery: beyond bowel perfusion
Salvador Morales-Conde, Sevilla, ES 17.15
Beyond TME – technique and results
of pelvic exenteration and sacrectomy
Paris Tekkis, London, UK
18.00
Get-Together with your colleagues 19.30
Industrial Exhibition FESTIVE EVENING

Information & Registration www.colorectalsurgery.eu